{"id":"W4391514274","doi":"10.1016/j.ebiom.2024.105006","title":"Decentralised, collaborative, and privacy-preserving machine learning for multi-hospital data","year":2024,"lang":"en","type":"preprint","venue":"EBioMedicine","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":30,"is_retracted":false,"has_abstract":true,"ca_institutions":"St. Michael's Hospital; Simon Fraser University; Vector Institute; University Health Network","funders":"Natural Sciences and Engineering Research Council of Canada; Defense Advanced Research Projects Agency; University of Toronto; Government of Canada; Vector Institute; Canadian Institute for Advanced Research; Government of Ontario; Intel Corporation; Alfred P. Sloan Foundation","keywords":"Computer science; Internet privacy; Patient privacy; Privacy policy; Information privacy; Computer security; Business; Political science; Health care; Law","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","metaepi_narrow","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.001435186,0.0006576883,0.0008129836,0.0005603089,0.0002448877,0.0006539148,0.05503263,0.000518111,0.00002009629],"category_scores_gemma":[0.04047521,0.0005766153,0.00007185321,0.0009760578,0.000392423,0.0005855843,0.5705138,0.001845165,0.00001874782],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001570141,"about_ca_system_score_gemma":0.0003466016,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001748031,"about_ca_topic_score_gemma":0.0000387896,"domain_scores_codex":[0.9951817,0.0001269673,0.0007166949,0.002567196,0.0006143875,0.0007930521],"domain_scores_gemma":[0.9807225,0.0004817853,0.0003955631,0.01791274,0.0002528586,0.0002345454],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004461124,0.0002513075,0.001744012,0.002870855,0.0006356181,0.0002619081,0.001334303,0.00003618795,0.0008031175,0.001402908,0.9284996,0.06211553],"study_design_scores_gemma":[0.001153556,0.0002085454,0.0002355453,0.0009807021,0.00009853239,0.00001612476,0.00008515064,0.8567467,0.0002874531,0.05967058,0.07991765,0.0005994],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.006080792,0.04109164,0.7554024,0.18197,0.006024154,0.002972327,0.00263908,0.003640432,0.0001791237],"genre_scores_gemma":[0.0349481,0.003047687,0.9574922,0.0002298832,0.0004706843,0.0001937102,0.003193133,0.0001084548,0.0003161224],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.8567106,"threshold_uncertainty_score":0.9996685,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06255268885975383,"score_gpt":0.3388253887363681,"score_spread":0.2762726998766142,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}